Sampling and Selection Bias in Research using Documented Skeletal Collections

Bias research in forensic anthropology—a field where practitioners are involved in multiple stages of an investigation and across a variety of analytical tasks—has focused primarily on the effects of cognitive bias on laboratory analyses. However, bias does not only arise from contextual information at the scene, but can occur as early as the initial research phase. However, to date only minimal research has focused on potential sources of bias within documented skeletal collections, despite their pivotal role in the production of analytical methods in the United States. Therefore, the purpose of this study is to identify potential sources of sample and selection bias in forensic anthropological research using documented skeletal collections. 

            The three goals of this project address different aspects of sample and selection bias potentially encountered during research development. The first goal is to use craniometric data from seven United States-based documented skeletal collections to investigate whether specific procurement strategies have resulted in collection specific sample bias. Identifying sample bias specific to each documented skeletal collection would have ramifications for the use of particular samples, previously published methods, and our understanding of human skeletal variation as it pertains to the biological profile (age, population affinity, sex, and stature). 

The second goal is to investigate whether collection samples are representative of the population of interest in forensic anthropology—forensic cases—by comparing craniometric data from one donated skeletal collection to craniometric data of individuals identified during forensic casework. The ultimate goal in forensic anthropology is to identify unknown decedents. If collection samples are not properly representative, established methods must be reexamined and countermeasures established to ensure appropriate samples are used going forward. 

The third goal is to use geospatial and geostatistical analyses to investigate whether the physical layout, storage of individuals, and curator involvement at documented human skeletal collections may introduce sample selection bias. Identifying potential biases in current sampling strategies is necessary to understand the extent of truly random samples in the discipline and to minimize such biases moving forward. 

            The culmination of this project will enable forensic experts to better understand the extent of bias introduced in the research process and its effect on the validity of historic methods still in use. Valid methods are required for expert court testimony and all attempts to mitigate bias are essential to upholding ethical standards in the criminal justice system. CA/NCF